核心概念
Zero-shot stylization in 3D scenes using text or visual input as conditioning factors is achieved through ConRF, outperforming existing methods.
統計
"Our experiment demonstrates that ConRF outperforms other existing methods for 3D scene and single-text stylization in terms of visual quality."
引用
"Our goal is to map the CLIP features space to the style space, simplifying the use of text or images as references to convey style."
"ConRF offers the capability to utilize either text or images as references, resulting in the generation of sequences with novel views enhanced by global or local stylization."